Orateur
Description
The ATLAS experiment at the Large Hadron Col-
lider (LHC) is operated at CERN and measures proton-proton
collisions at multi-TeV energies with a repetition frequency
of 40 MHz. Within the phase-II upgrade of the LHC, the
readout electronics of the liquid-argon (LAr) calorimeters
of ATLAS are being prepared for high luminosity operation
expecting a pileup of up to 200 simultaneous proton-proton
interactions. Moreover, the calorimeter signals of up to 25
subsequent collisions are overlapping, which increases the
difficulty of energy reconstruction by the calorimeter de-
tector. Real-time processing of digitized pulses sampled at
40 MHz is performed using field-programmable gate arrays
(FPGAs).
To cope with the signal pileup, new machine learning
approaches are explored: convolutional and recurrent neu-
ral networks outperform the optimal signal filter currently
used, both in assignment of the reconstructed energy to the
correct proton bunch crossing and in energy resolution. The
improvements concern in particular energies derived from
overlapping pulses.
Since the implementation of the neural networks targets
an FPGA, the number of parameters and the mathematical
operations need to be well controlled. The trained neural
network structures are converted into FPGA firmware us-
ing automated implementations in hardware description lan-
guage and high-level synthesis tools.
Very good agreement between neural network imple-
mentations in FPGA and software based calculations is ob-
served. The prototype implementations on an Intel Stratix 10 FPGA reach maximum operation frequencies of 344–
640 MHz. Applying time-division multiplexing allows the
processing of 390–576 calorimeter channels by one FPGA
for the most resource-efficient networks. Moreover, the la-
tency achieved is about 200 ns. These performance param-
eters show that a neural-network based energy reconstruc-
tion can be considered for the processing of the ATLAS LAr
calorimeter signals during the high-luminosity phase of the
LHC.
